In this project we have worked to strengthen the information management system of food safety and nutrition (SGI-SAN) in the Central america Dry Corridor to anticipate potential humanitarian crisis, identify priority territories to be attended and transfer the capacities to public institutions, universities and NGOs in Central America for the analysis of information, prediction of humanitarian needs and disclosure evidences about the situation.

This consultancy is part of the project "Maximising the impact of humanitarian assistance in the dry corridor by improving the generation and dissemination of Food Security and Nutrition data", where we proceeded to identify and characterise the Areas of Concern related to FNS Vulnerability using our own methodology already tested in validated research published in scientific journals of international relevance.

Some of the actions included in the project:

• Creation of a baseline model of SAN Vulnerability using Artificial Intelligence.
• Mapping, critical evaluation of technological solutions and elaboration of pre-agreements in different CA4 contexts.
• Experiencia piloto de pre-asistencia humanitaria basada en CVA como estrategia de preparación ante desastres en zonas de alto riesgo.
• Compilation of secondary databases related to the humanitarian field and elaboration of a baseline map of humanitarian vulnerability at the municipal level in CA4 under INFORM Index methodology.
• Elaboration of monthly analyses of humanitarian developments in CA4 with emphasis on agro-climatic aspects and for early warning purposes.
• Development of sample designs and primary data collection in areas of concern through telephone household surveys and dissemination of quarterly reports.
• Development of training manuals for expert humanitarian information managers (researchers) and users (NGOs, partners, local actors, etc.) of the HIS-CA4 and national and local workshops in disaster-affected areas.
• Construcción  un repositorio a modo de Data Warehouse con actualización mensual mediante fuentes secundarias, datos de teledetección a partir de sensores remotos y datos primarios de monitoreo. Para la obtención de los datos hemos implementado metodologías como Self Organizing Maps (SOM), Modelos Predictivos de Arboles de Decisión mediante Machine Learning, Análisis Agroclimáticos y Diseños Muestrales, todo ello reflejado en una plataforma digital con interfaz web operada por Power BI.

Project objectives:

• Information management for the anticipation of humanitarian crises caused by multiple causes in CA4 and the consensual identification of priority territories and populations to be addressed.
•  Improved capacity for anticipation, advocacy and coordinated response of local, national and regional humanitarian actors, with the support of academia and the private sector, to address disasters affecting CA4 territories exposed to multiple hazards.
• Improved information management for the anticipation of humanitarian crises caused by multiple causes in CA4 and the consensual identification of priority territories and populations to be addressed.
• Increased evidence-based empowerment and capacity of communities in CA4 exposed to high disaster risk to mobilise resources and manage preparedness and protection actions targeting their most vulnerable populations.

This entire process is reflected in the PREDISAN Central America online platform: